Antidiuretic response of a horse affected with pituitary pars intermedia dysfunction to desmopressin acetate


  • M. E. Moses,

    1. Department of Veterinary Medicine and Surgery, Veterinary Medical Teaching Hospital, College of Veterinary Medicine, University of Missouri, Columbia, Missouri, USA
    2. Steilacoom, Washington, USA
    Search for more papers by this author
  • P. J. Johnson,

    Corresponding author
    1. Department of Veterinary Medicine and Surgery, Veterinary Medical Teaching Hospital, College of Veterinary Medicine, University of Missouri, Columbia, Missouri, USA
    Search for more papers by this author
  • N. T. Messer,

    1. Department of Veterinary Medicine and Surgery, Veterinary Medical Teaching Hospital, College of Veterinary Medicine, University of Missouri, Columbia, Missouri, USA
    Search for more papers by this author
  • D. A. Wilson

    1. Department of Veterinary Medicine and Surgery, Veterinary Medical Teaching Hospital, College of Veterinary Medicine, University of Missouri, Columbia, Missouri, USA
    Search for more papers by this author



Central diabetes insipidus (DI) was diagnosed in a 20-year-old American Quarter Horse gelding that was concomitantly affected with pituitary pars intermedia dysfunction (PPID). The diagnosis of DI was supported by a positive response to administered desmopressin acetate. Diagnosis of PPID was supported by physical appearance and elevated plasma adrenocorticotropic hormone concentration following domperidone administration. The horse's physical condition improved following treatment with pergolide but long-term treatment with desmopressin was not undertaken and severe polyuria and polydipsia persisted. Desmopressin acetate appears to be useful for the diagnosis of DI in mature horses concomitantly affected with PPID.


Although polyuria and polydipsia (PUPD) are commonly reported clinical abnormalities in equine patients affected with pituitary pars intermedia dysfunction (PPID), the degree of increased voluntary water consumption in these patients is often mild or moderate. Severe PUPD may be implicative for diabetes insipidus (DI), a rarely reported condition in the equine species. It has been previously suggested that central DI may arise as a complication of PPID because clonal expansion of melanotropes in the intermediate lobe of the pituitary gland might lead to compression of the pars nervosa and reduced capacity of secretion of anti-diuretic hormone. In this clinical report, we describe a PPID-affected horse in which central DI was confirmed by the administration of desmopressin. However, it is likely that the development of central DI was significantly antecedent to that of PPID in this case.

Case history

A 20-year-old American Quarter Horse gelding was referred to the University of Missouri Veterinary Medical Teaching Hospital (VMTH) with a draining tract at the proximo-lateral aspect of the right femur of 2 weeks' duration. The draining tract had been treated with procaine penicillin G and trimethoprim-sulphadiazine. The horse had been regularly dewormed and vaccinated against tetanus, eastern and western equine encephalomyelitis, influenza and strangles. Mild diarrhoea and weight loss of 3 weeks' duration were present, in spite of a good appetite. The owner also reported that the horse drank copious volumes of water throughout the preceding 10 years.

On presentation, the horse was quiet, alert and responsive. The horse was thin with a body condition score of 3/9 and weighed 418 kg. The horse had a pendulous abdomen and long hair was present underneath the chin and on the caudal/palmar/plantar aspect of the thoracic and pelvic limbs. Faeces were matted on the caudal/plantar aspect of the pelvic limbs and a draining tract was present at the lateral aspect of the femur of the right pelvic limb. The rectal temperature was 37.8°C, the heart rate 40 beats/min and respiratory rate 24 breaths/min. A Grade 2/6 left-sided systolic heart murmur was evident with the point of maximal intensity at the heart base. Mucous membranes were normal. Loud borborygmi were evident throughout the abdomen. Mild lameness in the right pelvic limb was attributed to inflammation associated with the draining tract. During the course of the examination (within 60 min), the horse drank 32 l of water.

Following both digital palpation and ultrasonography, the draining tract was attributed to a penetrating skin wound with damage (superficial fragmentation) to the underlying third trochanter. It was considered likely that the wound resulted from a kick injury. Treatment for the draining tract included trimethoprim-sulphadiazine (15 mg/kg bwt per os q. 12 h) and daily flushing using a dilute povidone iodine solution. The horse was subsequently hospitalised in the VMTH for diagnostic tests pertaining to PUPD, weight loss and mild diarrhoea. The horse was fed a ration of ad libitum grass hay and a complete pelleted ration (Purina Equine Senior, 4 kg twice daily).

Haematology showed mild anaemia (6.24 × 109 red blood cells/l, reference range [rr] 6.8–12.9 × 109/l; PCV, 27%, rr 32–45%), mature neutrophilia (15.21 × 109/l, rr 2.26–8.58 × 109/l), and mild lymphopenia (1.35 × 109/l, rr 1.5–7.7 × 109/l). Plasma biochemistry showed hyperglycaemia (11.9 mmol/l, rr 4.6–6.3 mmol/l), hypoalbuminaemia (29 g/l, rr 35–44 g/l), hyperglobulinaemia (44 g/l, rr 24–41 g/l), slightly elevated aspartate aminotransferase (424 u/l, rr 175–394 u/l), and slightly elevated creatine kinase (1468 u/l, rr 109–456 u/l). Abnormalities identified in a urine sample collected (free-catch) on admission included a low specific gravity (USG; 1.001), aciduria (urinary pH 6.5) and glycosuria (13.9 mmol/l, rr 0 mmol/l).

It was considered likely that, based on the horse's age and results of physical examination, weight loss and PUPD might be attributable to pituitary pars intermedia dysfunction (PPID). Haematological changes in this horse were characteristic for either a stress response or PPID. Anaemia may have resulted from chronic inflammation associated with the draining tract or other unidentified inflammatory changes in the body. Hyperglycaemia and glycosuria probably resulted from either stress or PPID but diabetes mellitus was also a possibility. Hypoalbuminaemia and hyperglobulinaemia may have resulted from a chronic inflammatory process, such as that associated with the wound. Hypoalbuminaemia might have also resulted from underlying protein-losing colonopathy (associated with diarrhoea), a small intestinal malabsorptive condition or kidney disease (PUPD). However, kidney disease was considered unlikely because analysis of urine failed to identify proteinuria. Elevated plasma muscle enzyme activities were attributed to traumatic muscle damage in the proximal aspect of the injured right pelvic limb. That the horse was producing large quantities of hyposthenuric urine in the face of exaggerated voluntary water consumption was implicative for diabetes insipidus (DI), possibly resulting from PPID. Other causes for PUPD that were considered included renal disease, diabetes mellitus, psychogenic water consumption, medullary washout, excessive salt consumption and hyperthyroidism.

Further diagnostic tests for PUPD included quantification of water consumption, repeated determination of plasma glucose concentration, bacteriological culturing of urine, repeated urinalyses, and measurement of endogenous plasma ACTH before and after treatment using domperidone (2.5 mg/kg bwt) (Sojka et al. 2006). Mean voluntary water consumption, measured over the course of 3 days, was determined to be 215 ml/kg bwt/day (rr 30–60 ml/kg bwt/day) (Schott 2006). Although it was not quantified, the volume of urine output was also clearly excessive.

Repeated determinations of plasma glucose concentration revealed that hyperglycaemia was not present following admission to the VMTH, eliminating consideration for diabetes mellitus and implying that transport stress probably accounted for this abnormality on presentation. Therefore, hyperglycaemia was ruled out as the cause for PUPD. Bacteriological culturing of urine collected aseptically through a catheter yielded negative results (noting that the horse was being treated with broad-spectrum antimicrobials). Results of repeated urinalyses were normal (including microscopic examinations of the sediment) with the exception of hyposthenuria (USG ranged from 1.001–1.004); glycosuria was not evident after admission.

Endogenous plasma adrenocorticotropic hormone (ACTH) measured at 08.00 h on Day 2 of hospitalisation was 22.6 ng/l (rr 9–35 ng/l). Four hours following domperidone treatment, the endogenous plasma ACTH concentration was 89.4 ng/l. That the ACTH concentration had increased by a factor of 4 was indicative of PPID (Sojka et al. 2006; Miller et al. 2008).

Diagnostic tests for diarrhoea included bacteriological culturing of faeces for Salmonella on 5 different days, faecal ELISA for Clostridium difficile exotoxins, parasitological examination of faeces, abdominocentesis, per rectum abdominal palpation and abdominal ultrasonography. Results of bacteriological studies yielded negative results and the faeces were negative for parasites. Abdominocentesis yielded normal peritoneal fluid (total nucleated cell count, 0.45 × 109 cells/l, rr <5 × 109 cells/l; protein concentration, <25 g/l, rr <25 g/l). Both per rectum abdominal palpation and an abdominal ultrasonographic examination failed to identify any abnormalities. During the hospitalised period, the diarrhoea resolved within 24 h.

A water deprivation study was initiated on the third day of hospitalisation (Table 1). The horse's weight was not monitored during the test because diarrhoea required accommodation in isolation. Therefore, the horse was held without access to food or water for 24 h, during which time hydration status, heart rate, PCV, plasma total solids, USG, plasma creatinine concentration and subjective evaluation of urinary output were assessed on a regular basis. After 24 h, the USG was 1.010 and water deprivation was discontinued. The fact that the horse did not produce concentrated urine (USG>1.020) following water deprivation tended to rule out psychogenic factors as the cause of PUPD. The horse developed dry mucous membranes during the course of the water deprivation test and both the PCV and plasma total solids had increased (PCV increased from 29 to 38% and plasma total solids increased from 66 to 78 g/l) during the test, suggesting that the horse had experienced a significant stimulus to release arginine vasopressin. The horse remained bright and alert throughout the period of water deprivation.

Table 1. Results of quantified water consumption, urine specific gravity measurements, and determinations of urine and plasma osmolality over the course of hospitalisation
DayQVWC (ml/kg bwt/24 h)Urine specific gravityUosm (mOsmol/kg)Posm (mOsmol/kg)
  1. QVWC: quantified voluntary water consumption; Uosm: urine osmolality; Posm: plasma osmolality.

30Time 0: 1.000
Water deprivation test (20 h)Time +8 h: 1.010
Time +20 h: 1.010
565Time 0: 1.00281280
Desmopressin administration (32 h)Time +7 h: 1.01496281
Time +23 h: 1.020546282
6235Time +30 h: 1.006186282
Time +31 h: 1.005122280
Time +32 h: 1.004110280

On the fifth day, urine and plasma were collected for determination of osmolality and the horse was treated using desmopressin acetate (20 µg, i.v.) (Barnes et al. 2002). Desmopressin acetate (1-desamino-8-d-arginine vasopressin) is a powerful synthetic analogue of arginine vasopressin (anti-diuretic hormone). Following treatment, voluntary water consumption was measured and urine output was assessed subjectively throughout the subsequent 32 h (Table 1). Throughout the treatment period, plasma osmolality (Posm) was measured and urine was collected for measurement of USG and urine osmolality (Uosm). Desmopressin treatment caused the water consumption to decrease to 65 ml/kg bwt/24 h and urination was observed only 3 times during the 24 h post treatment period. By 7 h following treatment, the USG had increased to 1.014. At 23 h, the USG was 1.020. By 32 h post treatment, water consumption had increased and the USG had decreased to 1.004. Prior to treatment with desmopressin, the Posm was 280 mOsm/kg (Posm and Uosm were measured by freezing point depression) and the Uosm was 81 mOsm/kg. At 23 h following treatment, Uosm had increased to 546 mOsm/kg while Posm remained at 282 mOsm/kg. By 32 h post treatment, Uosm had decreased to 110 mOsm/kg.

In conclusion, treatment with desmopressin restored the horse's ability to concentrate the urine. We concluded that PUPD was primarily attributable to central DI and that pituitary pars nervosa compression may have resulted from clonal expansion of melanotropes in the pars intermedia. The gelding was discharged from the VMTH for further treatment at home using trimethoprim-sulpha, wound irrigation and pergolide mesylate (1.0 mg per os q. 24 h) (Donaldson et al. 2002; Perkins et al. 2002).

The wound on the right pelvic limb healed quickly and the horse's body condition improved significantly following institution of pergolide treatment and dietary adjustments. Further signs of diarrhoea were not observed following return home. However, the owner reported that PUPD was unaffected by the pergolide treatment after 10 months. The horse died at pasture within 12 months of discharge from the VMTH and a post mortem examination was not performed. The owner reported that the horse had continued to gain weight following treatment with pergolide.


In this horse, a single treatment with desmopressin resulted in a marked reduction in voluntary water consumption, reduced urination and increased Uosm. These observations suggest that PUPD was probably attributable to compromised production and release of anti-diuretic hormone (ADH).

It should be noted that the highest attained Uosm that was observed in this horse (546 mOsm/kg) was not substantial and represented a Uosm:Posm ratio of only 1.94:1. As a general rule, normal urinary concentrating ability is characterised by a Uosm:Posm ratio of >2. Commonly, the Uosm in healthy mature horses is 900–1200 mOsm/kg (representing a value 3–4 times that of the plasma) (Brashier 2006). Possibly, the osmolality of this horse's renal medullary interstitium had been somewhat ‘washed out’ by the effects of chronic PUPD.

Although it is tempting to explain the development of severe PUPD as a complication of PPID causing reduced ADH release by virtue of physical compression of the pars nervosa by an expanded PI, severe PUPD had been present in this horse since it was aged 10 years, well prior to the onset of signs of PPID. It seems likely that DI may have been present before the development of PPID (and was therefore unrelated to PPID). Although PPID is most commonly diagnosed in horses and ponies aged >18 years, it is occasionally reported in younger animals (Donaldson et al. 2004; Messer and Johnson 2007). Moreover, although treatment using pergolide and an improved diet led to clinical improvement in terms of body condition and diarrhoea, the PUPD was unchanged, further suggesting that DI might have been independent of PPID in this case.

Polyuria and polydipsia have sometimes been reported to occur in horses and ponies affected with PPID (Coffman 1981; Schott 2002; Messer and Johnson 2007). However, in most PPID-affected horses, the increased appetite for water is less pronounced than it was in the horse of this report in which central DI was also diagnosed. Alternative explanations for PUPD in PPID-affected horses include: solute diuresis resulting from hyperglycaemia and glucosuria; ascending urinary tract infection resulting from immunocompromise; cortisol-mediated inhibition of ADH effect in the distal nephron; and central DI resulting from clonal expansion of melanotropes in the PI leading to physical encroachment on the neurohypophysis (pars nervosa or the hypothalamus) (Schott 2002; Messer and Johnson 2007). Unlike horses, central DI has been previously documented as an important component of pituitary neoplasia in dogs (Harb et al. 1996).

Polyuria and polydipsia represent an infrequently encountered clinical problem for veterinarians working in the equine field (Coffman 1981). Differential diagnoses for PUPD include psychogenic water consumption, renal failure and PPID. Less commonly, PUPD may be attributed to diabetes mellitus, hyperthyroidism, endotoxaemia, the ingestion or administration of excessive sodium chloride, veterinary therapeutic strategies (such as administration of diuretics, corticosteroids, and fluid therapy), renal helminthiasis, and DI (Buntain and Coffman 1981; Coffman 1981; Sturgeon and Bassett 2000; McKenzie 2007).

Diabetes insipidus results from insufficient or absent ADH effect at the epithelial cells of the cortical and medullary collecting ducts (Robertson 1995; Rose and Post 2001; Jane et al. 2006). Nephrogenic DI is a congenital or acquired disorder in which neurohypophyseal function and ADH release are normal but in which renal responsiveness to appropriate plasma concentrations of ADH is reduced (Rose and Post 2001; Jane et al. 2006). Acquired causes of nephrogenic DI include acute and chronic renal diseases, genetic factors, hypokalaemia, hypercalcaemia, various drugs and medullary washout (loss of hyperosmolal renal medullary interstitium as a result of protracted psychogenic polydipsia) (Rose and Post 2001).

Central DI is more common in human patients and results from a failure of ADH production or release from the neurohypophysis (development of PUPD implies that there has been an 80–90% reduction of the secretory output of ADH by the neurohypophysis) (Jane et al. 2006). In addition to hereditary forms of the condition, acquired central DI may result from head trauma, hypoxic or ischaemic encephalopathy, encephalitis, vascular anomalies, neoplasia, or familial and idiopathic factors (Robertson 1995; Rose and Post 2001; Jane et al. 2006). Although central DI has been reported to occur as a result of pituitary neoplasia, it is much more commonly attributable to hypothalamic causes in human patients (Jane et al. 2006).

Both forms of DI are rare in horses and ponies (Chenault 1969; Filar et al. 1971; Satish and Sastry 1978; Breukink et al. 1983; Vente and Wijsmuller 1983; Brashier 2006). Although some reports were likely attributable to psychogenic PD, there are at least 3 well-documented reports of nephrogenic DI (Schott et al. 1993; Brashier 2006) and 2 well-documented reports of central DI in mature horses (Filar et al. 1971; Breukink et al. 1983). One case of congenital central diabetes insipidus has also been reported in a 10-day-old Friesian colt (Kranenburg et al. 2010). Nephrogenic DI has been reported in 2 related Thoroughbred colts suggesting that it may be inherited as an X-linked disorder, as is the case in man (Schott et al. 1993). Central DI has been previously reported as a component of encephalitis in one horse and as an idiopathic syndrome in another (Filar et al. 1971; Breukink et al. 1983).

In 2 cases of nephrogenic DI, it had been possible to demonstrate high resting plasma arginine vasopressin (ADH) concentration and an elevation in the circulating plasma concentration of arginine vasopressin during a water deprivation challenge (Schott et al. 1993; Brashier 2006). In another case, a failure to demonstrate an elevation in the plasma arginine vasopressin concentration during water deprivation supported a diagnosis of central DI (Vente and Wijsmuller 1983). We did not measure the plasma arginine vasopressin concentration in this case because we did not identify a clinical laboratory that offered the assay. The dose of desmopressin acetate, a synthetic arginine vasopressin (ADH) analogue, that we used had been previously shown to be sufficient to cause production of concentrated urine in normal horses and in horses affected with nephrogenic DI (Breukink et al. 1983; Barnes et al. 2002; Brashier 2006).

A specific cause for the mild diarrhoea was not identified in this case although several commonly recognised aetiologies including salmonellosis, Clostridium difficile infection and endoparasitism were tentatively ruled out. Diarrhoea resolved following institution of pergolide treatment suggesting that colonic floral imbalance might have been present as a result of immunocompromise. Human patients with DI tend to develop constipation in the face of prodigious water consumption (Jane et al. 2006). Diarrhoea resolved in the face of ongoing severe polydipsia following discharge from the VMTH in our case, suggesting that it may not have been a result of increased water intake.

Although there have been 2 previous reports of central DI in horses, it has not been previously described in PPID-affected horses or ponies (Filar et al. 1971; Breukink et al. 1983). Although PUPD is sometimes observed in PPID-affected equids, a specific explanation for PUPD in these cases is usually not defined (Schott 2002; Messer and Johnson 2007). Recent work has demonstrated that PPID is a primary hypothalamic condition in which degeneration of dopaminergic nerves (cell bodies in the periventricular nucleus of the hypothalamus) resulting from oxidative stress leads to loss of physiological inhibitory influence in the PI (McFarlane et al. 2005). Although the cause of central DI in horses has not been elucidated, one might speculate that neurodegenerative changes in the neurons originating in the supraoptic and paraventricular nuclei of the hypothalamus could similarly account for failed production of arginine vasopressin (ADH) and its release from the pars nervosa.

Domperidone has been advocated as an evocative agent for use in endocrinological testing for PPID because the resting (baseline) plasma ACTH concentration may be normal, even in those horses with Grade 3 or Grade 4 PI lesions. The dose of domperidone that we used was based on an earlier recommendation by Sojka et al. (2006). More recently, the same authors have proposed that a higher dose for domperidone (3.3 mg/kg bwt) should be used during this diagnostic test protocol (Miller et al. 2008). Similarly, we measured the plasma ACTH concentration at only the zero time and at +4 h following administration of domperidone based on the original recommendation (Sojka et al. 2006). Miller et al. (2008) subsequently reported a greater correlation between ACTH and histological grade at +8 h after domperidone (compared to the +4 h sample time). More recently, the use of thyrotropin-releasing hormone as an evocative agent (measuring ACTH before and after administration of thyrotropin-releasing hormone) was shown to be superior to domperidone for the diagnosis of PPID in horses (Beech et al. 2011).

In human medicine, lifelong daily vasopressin injections are required for the management of central DI. Injectable vasopressin is cost prohibitive for systemic administration in horses at present. Possibly, inhalant desmopressin or desmopressin eye drops could be considered for the management of central DI in horses in the future. Desmopressin eye drops represent a noninvasive, practical and effective way of hormone replacement treatment for DI in horses as well as human patients (Kranenburg et al. 2010). While treatment using desmopressin acetate may not be medically crucial in all cases, it may be helpful for purposes of improving the quality of life for those horses affected with both central DI and PPID.

Authors' declaration of interests

No conflicts of interest have been declared.